The anticipated demand for DNA-based genetic information, in both clinical and basic research laboratories, is virtually unlimited. The biochemical manipulations for DNA analysis are well characterized but have not been assembled into a simple automated system. This continuation proposal will develop a fully integrated technology for DNA processing and analysis. All the steps for extracting genotype information by the polymerase chain reaction (PCR) will occur within a single self-contained system. Both the biochemical and electronic components of the system are being fabricated on silicon or silicon/glass substrates, yielding a micromechanical integrated DNA analysis technology (MIDAT). An enormous range of mechanical and electrical devices can be constructed on silicon using advanced micromechanical fabrication techniques. Since the characteristics of silicon are well understood, designs for components can be rapidly generated, tested, and replicated. The low marginal cost of mass-produced silicon devices can allow the system to be disposable, thereby reducing the risk of sample cross-contamination and error. By using a reduced quantity of reaction components and a single-use liquid handling system, microscale genotyping will result in significantly lowered sample processing costs and increased sample throughput. This proposal has three specific goals: 1) Complete an integrated DNA analysis device at the 100 nl reaction volume scale. The devices will be used for conventional, single-step sequence-tagged-site PCR amplification and product size analysis. The devices will provide a robust and consistent testing platform for further improvements in materials, channel designs, surfaces, and sensors. 2) Complete an integrated DNA analysis device at the 10 nl reaction volume scale. The system will be compatible with basic PCR genotyping methods and generate over 100 devices per fabricated wafer. This size scale will provide a laboratory-grade genotyping system that is competitive in cost and efficiency with many current macro-scale methods. 3) Development of functions and parallel processes for laboratory genetic testing. Components will be developed to conform with the demands of more complex genotyping methods. An emphasis will be placed on early steps in sample processing and on parallel analysis. The number of parallel PCR processing units per wafer will be maximized and circuitry developed for process control.